MR Fingerprinting (MRF)

MR Fingerprinting (MRF)

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

# Method Score PSNR (dB) SSIM Source
🥇 MRF-Former 0.773 33.5 0.930 ✓ Certified MRF transformer, 2024
🥈 MRF-Net 0.723 31.5 0.895 ✓ Certified Cohen et al., Med. Phys. 2018
🥉 MANTIS 0.595 27.0 0.790 ✓ Certified Cohen et al., MRM 2018
4 SVD-MRF 0.467 23.5 0.650 ✓ Certified Ma et al., Nature 2013

Dataset: PWM Benchmark (4 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

# Method Overall Score Public
PSNR / SSIM
 Dev
PSNR / SSIM
 Hidden
PSNR / SSIM
 Trust Source
🥇 MRF-Former + gradient 0.718
0.785
32.3 dB / 0.946
0.703
28.6 dB / 0.893
0.667
26.67 dB / 0.850
✓ Certified MRF tissue quantification transformer, 2024
🥈 MANTIS + gradient 0.609
0.643
24.94 dB / 0.800
0.612
24.17 dB / 0.775
0.572
22.91 dB / 0.728
✓ Certified Cohen et al., MRM 2018
🥉 MRF-Net + gradient 0.576
0.758
30.46 dB / 0.924
0.519
20.19 dB / 0.608
0.450
18.46 dB / 0.523
✓ Certified Cohen et al., Med. Phys. 2018
4 SVD-MRF + gradient 0.516
0.542
20.76 dB / 0.635
0.516
20.45 dB / 0.620
0.489
19.5 dB / 0.574
✓ Certified Ma et al., Nature 2013

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →
Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%
Public 3 scenes

Full-access development tier with all data visible.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.

How to use: Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.

What to submit: Reconstructed signals (x_hat) and corrected spec as HDF5.

Public Leaderboard
# Method Score PSNR SSIM
1 MRF-Former + gradient 0.785 32.3 0.946
2 MRF-Net + gradient 0.758 30.46 0.924
3 MANTIS + gradient 0.643 24.94 0.8
4 SVD-MRF + gradient 0.542 20.76 0.635
Spec Ranges (3 parameters)
Parameter Min Max Unit
dictionary_resolution_(t1,_t2) -1.0 2.0 -
b1_inhomogeneity -3.0 6.0 -
undersampling_artifact -4.0 8.0 -
View Details →
Dev 3 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), and spec ranges only.

How to use: Apply your pipeline from the Public tier. Use consistency as self-check.

What to submit: Reconstructed signals and corrected spec. Scored server-side.

Dev Leaderboard
# Method Score PSNR SSIM
1 MRF-Former + gradient 0.703 28.6 0.893
2 MANTIS + gradient 0.612 24.17 0.775
3 MRF-Net + gradient 0.519 20.19 0.608
4 SVD-MRF + gradient 0.516 20.45 0.62
Spec Ranges (3 parameters)
Parameter Min Max Unit
dictionary_resolution_(t1,_t2) -1.2 1.8 -
b1_inhomogeneity -3.6 5.4 -
undersampling_artifact -4.8 7.2 -
Submit Reconstruction View Details →
Hidden 3 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

What you get: No data downloadable. Algorithm runs server-side on hidden measurements.

How to use: Package algorithm as Docker container / Python script. Submit via link.

What to submit: Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Hidden Leaderboard
# Method Score PSNR SSIM
1 MRF-Former + gradient 0.667 26.67 0.85
2 MANTIS + gradient 0.572 22.91 0.728
3 SVD-MRF + gradient 0.489 19.5 0.574
4 MRF-Net + gradient 0.450 18.46 0.523
Spec Ranges (3 parameters)
Parameter Min Max Unit
dictionary_resolution_(t1,_t2) -0.7 2.3 -
b1_inhomogeneity -2.1 6.9 -
undersampling_artifact -2.8 9.2 -
Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

M → F → S → D

M Modulation
F Fourier
S Sampling
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
d_r dictionary_resolution_(t1,_t2) Dictionary resolution (T1, T2) (-) 0.0 1.0
b_i b1_inhomogeneity B1 inhomogeneity (-) 0.0 3.0
u_a undersampling_artifact Undersampling artifact (-) 0.0 4.0

Credits System

40%
Platform Profit Pool
Revenue allocated to benchmark rewards
30%
Winner Share
Top algorithm receives from pool
$100
Min Withdrawal
Minimum payout threshold
Spec Primitives Reference (11 primitives)
P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.